Holocentridae

Abstract: Genetic surveys of reef fishes have revealed high population connectivity within ocean basins, consistent with the assumption that pelagic larvae disperse long distances by oceanic currents. However, several recent studies have demonstrated that larval retention and self-recruitment may be higher than previously expected. To assess connectivity in tropical reef fishes, we contribute range-wide mtDNA surveys of two Atlantic squirrelfishes (family Holocentridae). The blackbar soldierfish, Myripristis jacobus, has a pelagic juvenile phase of about 58 days, compared to about 71 days (~22% longer) in the longjaw squirrelfish, Holocentrus ascensionis. If the pelagic duration is guiding dispersal ability, M. jacobus should have greater population genetic structure than H. ascensionis. In comparisons of mtDNA cytochrome b sequences from 69 M. jacobus (744 bp) and 101 H. ascensionis (769 bp), both species exhibited a large number of closely related haplotypes (h=0.781 and 0.974, π=0.003 and 0.006, respectively), indicating late Pleistocene coalescence of mtDNA lineages. Contrary to the prediction based on pelagic duration, M. jacobus has much less population structure (φST=0.008, P=0.228) than H. ascensionis (φST=0.091, P<0.001). Significant population partitions in H. ascensionis were observed between eastern, central and western Atlantic, and between Brazil and the Caribbean in the western Atlantic. These results, in combination with the findings from 13 codistributed species, indicate that pelagic larval duration is a poor predictor of population genetic structure in Atlantic reef fishes. A key to understanding this disparity may be the evolutionary depth among corresponding taxonomic groups of “reef fishes”, which extends back to the mid-Cretaceous and encompasses enormous diversity in ecology and life history. We should not expect a simple relationship between pelagic larval duration and genetic connectivity, among lineages that diverged 50–100 million years ago.

Abstract: 1. Auditory sensitivity as a function of frequency has been behaviorally determined for two species of fish from the teleost family Holocentridae which is characterized by marked variation in peripheral auditory structures. 2. Best sensitivity measured forMyripristis kuntee was -50 dB re: 1 dyne/cm2 for frequencies between 300 and 2,000 Hz, while best sensitivity measured forAdioryx xantherythrus was -28 dB at 500 Hz (Figs. 1 and 2). 3. Both species can detect sounds at 100 Hz while the high frequency end of the auditory range extends up to 3,000 Hz forM. kuntee and to 800 Hz forA. xantherythrus (Figs. 1 and 2). 4. It is hypothesized that these differences in auditory capabilities are related to differences in sound transmission characteristics of the peripheral auditory system.

Abstract: To increase our understanding of how large numbers of similar species of reef fishes coexist, we have determined the components of feeding-niche separation among the members of a feeding guild of coral reef fishes. The seven West Indian species of shallow-water squirrelfishes (Holocentridae) comprised >99W of the nocturnally active, benthic-crustacean-feeding fishes at five sites off St. Croix, Virgin Islands. Resource utilization frequencies were determined for food and foraging habitat. Food was partitioned by taxon between the four species that consumed predomi- nantly shrimps and the three species that ate mainly crabs. Food was secondarily partitioned by body size of prey items, particularly in the principal food category. Differences in foraging microhabitat (position within a reef zone) were as important as food differences in separating species. In all cases but one, overall feeding niche overlap was =-.25 or less, with a mean value of .13. This low value contradicts statements that reef fishes are generalists with broadly overlapping resource utilization and, compared with overlap values in other guilds, suggests that the mechanisms permitting great local diversity within coral reef fish guilds are not basically different from those operating in other ecosystems. The mean overlap in food among the 6 commonest species was significantly different from the mean overlaps of 50 randomly generated competition-free communities (Sale 1974), sug- gesting that interspecific competition has played a role in the evolution of this assemblage.